Rethinking the Stability-Plasticity Trade-off in Continual Learning from an Architectural Perspective
This work addresses the challenge of enabling neural networks to learn incrementally without forgetting, which is crucial for real-world AI applications, though it is incremental as it builds on existing continual learning methods.
The paper tackles the stability-plasticity trade-off in continual learning by analyzing network architecture, showing that deeper networks improve plasticity and wider networks enhance stability under equal parameters, and introduces Dual-Arch, a plug-in framework that boosts existing methods' performance while reducing parameters by up to 87%.
The quest for Continual Learning (CL) seeks to empower neural networks with the ability to learn and adapt incrementally. Central to this pursuit is addressing the stability-plasticity dilemma, which involves striking a balance between two conflicting objectives: preserving previously learned knowledge and acquiring new knowledge. While numerous CL methods aim to achieve this trade-off, they often overlook the impact of network architecture on stability and plasticity, restricting the trade-off to the parameter level. In this paper, we delve into the conflict between stability and plasticity at the architectural level. We reveal that under an equal parameter constraint, deeper networks exhibit better plasticity, while wider networks are characterized by superior stability. To address this architectural-level dilemma, we introduce a novel framework denoted Dual-Arch, which serves as a plug-in component for CL. This framework leverages the complementary strengths of two distinct and independent networks: one dedicated to plasticity and the other to stability. Each network is designed with a specialized and lightweight architecture, tailored to its respective objective. Extensive experiments demonstrate that Dual-Arch enhances the performance of existing CL methods while being up to 87% more compact in terms of parameters. Code: https://github.com/byyx666/Dual-Arch.